PDA

View Full Version : Serial logging and tuning with 2 LC-1's



truethinker
June 4th, 2012, 01:47 PM
I'm sure I read a post on this a while ago but i can't find it. . . I just got done installing two LC-1 units(one for each exhaust bank) in my 2003 Sierra Denali.

I have them daisy chained together and connected through a taquickness null modem cable. Is there a way for me to log and AutoVE with both? I'm thinking log them as an average?

Thanks
Jesse

truethinker
June 5th, 2012, 01:07 AM
Nevermind, I think I got it worked out in calc.pids

joecar
June 5th, 2012, 01:05 PM
Post your calc_pids.txt file, let me sanity check it.

truethinker
June 6th, 2012, 01:04 AM
Thanks joecar. I've been away from tuning and EFIlive for a few years and I was new to it then.

13316

truethinker
June 6th, 2012, 09:16 AM
Just tried out the new LC-1 setup and everything seems to be running well in place of the old NB02's. I think I got the PID setup for averaging the two LC-1's but I'm getting no data and this is what it says in "more info"

PID value cannot be determined because
the following error would occur:
List index out of bounds (-1)

*CLC-00-020
AFR 10.0 20.0 .3 "({EXT.WO2AFR1}+{EXT.WO2AFR2})/2



CALC.EXTWO2AFRAV F020 CLC-00-020 AFR WO2-Serial "AFR from Serial WB - 1/2 Average"

joecar
June 6th, 2012, 12:15 PM
Just tried out the new LC-1 setup and everything seems to be running well in place of the old NB02's. I think I got the PID setup for averaging the two LC-1's but I'm getting no data and this is what it says in "more info"

PID value cannot be determined because
the following error would occur:
List index out of bounds (-1)

*CLC-00-020
AFR 10.0 20.0 .3 "({EXT.WO2AFR1}+{EXT.WO2AFR2})/2



CALC.EXTWO2AFRAV F020 CLC-00-020 AFR WO2-Serial "AFR from Serial WB - 1/2 Average"On the end of that line you forgot to place a closing quote.

Same on the end of CLC-00-021.

joecar
June 6th, 2012, 12:17 PM
Do you use any of the analog wideband stuff...?

joecar
June 6th, 2012, 12:20 PM
Which pids are you using...?

joecar
June 6th, 2012, 12:23 PM
Also, the calc_pids.txt file needs two blank lines at its end.

joecar
June 6th, 2012, 12:48 PM
Also, some of the pids were being calculated out-of-order...

EFILive calculates the pids in CLC-00-xxx numerical order (i.e. lowest xxx first)...

some of the pids were being used before they were calculated (i.e. their previous frame's value was used in the current frame's calculation).


I edited your file a little, see attached, please try it and let me know.

:)

truethinker
June 6th, 2012, 01:39 PM
Thanks a lot! I just got done modifying my original file per your previous suggestions and was about to go out to the truck when I got the email notice that you had replied. Glad I didn't waste my time with the out of order PID's.


The analog pid's are for a '96 chevy 1500 that was my first fuel injected project/daily driver. I did the 411 and 4L80E swaps along with a ramjet 350 intake, mid-performance cam and heavily ported/polished heads. The thing idles horribly and dumps fuel out the exhaust so needless to say, it needs tuned. From what I've been reading, tuning for heads and cam is no small task. I'm trying to get my feet wet with legitimate tuning on my '03 so that I can get the '96 running better as I gave it to my brother who is going to be a sophomore and the college allows everyone but freshman to have vehicles on campus. The problem is he plays football for them and due to this, his summer is very short.

I pretty much have 3 weeks to get both trucks tuned as I am driving mine to Northern Maine mid-summer and would love to see some better fuel mileage and driveability.

Thanks again for your help. I've attached a screenshot of my pid's list per your request. I also, included my calc_pid file as I made a slight change. I changed CLC-00-300 to reflect the pid in the CALC VET turtorial.

One other thing, in looking at my file, I have to admit that I'm not even sure why I have *CLC-00-273, *CLC-00-430 or *CLC-00-435. I'm sure there was a reason I put them in there but who knows as I don't remember. I don't see what they would be used for with my current set-up.


Thanks

joecar
June 6th, 2012, 01:56 PM
I had used the pid {CALC.DAT.K} as a concise shortcut for ({GM.DYNAIRTMP_DMA.C}+273.15) and its chart label is easier to see... but either can be used and both are correct

(actually I updated the Calc.VET tutorial thread with the newer calc_pids.txt which uses CALC.DAT.K/CLC-00-273).

Those other pids are used for comparing the various dynamic air masses.

truethinker
June 6th, 2012, 02:17 PM
ooh okay, so that's where I got it. And then I updated it from the PDF version which is written the other way. In that case I'm going to change back to the original way I had it.

Alright, so I think I have it all nailed down. I'll log my way to work in the morning and see how it turns out.

joecar
June 7th, 2012, 03:43 AM
Yes, the pdf hasn't yet caught up. sorry.

truethinker
June 7th, 2012, 04:28 AM
No worries! I got everything hooked up and logging today on the way to work. Seems to have worked well. I was a little confused for a moment when I first looked at the map data and say a bunch of regular percentages. I knew that's what would happen but some part of my brain was still stuck on regular autoVE and I was expecting to see a bunch of 1.1's and 0.92's and so on. haha I'll load my new tune and log on my way home from work tonight.

On a side note,(and I know this isn't related to the title of this thread) do I need to adjust my setting to account for the fuel around here being 10% ethanol? I've been reading a bunch of posts about that and just trying to see what, if any, changes are really necessary. I'm mostly concerned because I intend to try activate and fine tune lean cruise. My end goal is maximum fuel mileage possible.

I think I have a decent of understanding of what gets changed and how. But my question is simply, do I need to and will it help or hurt mileage. My "B3601" appears to be set-up with the ability to deal with flex-fuel but it is flat lined at 14.63. My truck is an '03 and is not flex-fuel(and "B3101" is set to "no").

joecar
June 7th, 2012, 04:56 AM
If you're running E10, set B3601 to 14.2 in your tune.

joecar
June 7th, 2012, 04:58 AM
Relationship between AutoVE/AutoMAF and Calc.VET/Calc.MAFT, see post #29 here: Summary-Notes (http://forum.efilive.com/showthread.php?14188-Summary-Notes)

truethinker
June 7th, 2012, 07:18 AM
Very interesting reading. Thanks again for your help

truethinker
June 7th, 2012, 01:46 PM
So I logged my drive home from work, making sure I included some highway, mountain and in-town driving. I tried to keep my throttle as steady as possible and used lower gears as opposed to braking. Upon comparing the log to the OEM VE table, my log look sadly low. I didn't expect to see my VE that much lower than Factory.



Can you take a look at this and see what you think? Is this a logical log or do you feel I might have something wrong?

See attached original VE and Logged File

I entered the PE value incorrectly so the filter removed anything over 1.06lam. (Yea, I went lean in PE rather than rich but I fixed it with no harm done.)

1332713326

joecar
June 7th, 2012, 02:45 PM
Post your tune file, log file and the calc_pids.txt you used (unless you used the one from post #10 above).

truethinker
June 7th, 2012, 03:16 PM
13330

13331

13332

joecar
June 8th, 2012, 02:04 PM
13330

13331

13332No good (very bad)... you're commanding stoich AFR at WOT, see frames 15481-15654...

something is really wrong and I can't see what...

set these as follows:
- B3643 to 1.0000,
- B3644 to 0.00625,
- B3645 to 0.00625.

truethinker
June 8th, 2012, 02:15 PM
I know why, I actually realized that I made a pretty drastic mistake shortly after I got home from that drive. . . .

I read the bellow pasted line in the VET Summary Notes and entered the EQ value into the table referenced. . . not noticing that it was in lambda and as such I was commanding a lean condition.

"5. B3618: PE Modifier Based on Rpm (EQ): 1.16
Safe adequate Fueling regardless of Fuel Type"

That mistake is fixed in the tune file I posted.

joecar
June 8th, 2012, 02:21 PM
Ah, ok, I couldn't see the reason why PE wasn't being commanded (so I thought the PE ramp rates were wrong).

truethinker
June 8th, 2012, 02:28 PM
Sorry about that, I should have noted that in the post. Aside from the obvious PE issues, is it normal for all of my VE percentages to be reading that much below stock? Just seems strange to me but then I have no real past experience to rely upon.

joecar
June 8th, 2012, 02:41 PM
Hi Jesse,

It means that the MAF table is not correct.


( so your Calc.VET effort will do this:
- correct the MAF using SELBEN;
- calculate a new VE based on the corrected MAF )

truethinker
June 9th, 2012, 09:07 AM
I think I got a better log today and definitely got a good adjustment on my MAF.

I saw a couple of issues though.


In this screenshot you can see a spike in my LC-1 Lambda reading. Just curious what that might be.
13339


The other issue I noticed is at WOT. The computer commands an EQ of 1.16 as suggested but the WBO2 shows that I go more and more lean the higher the RPMs go. Watch between 23757 and 23900. Could this just be the fact that the truck is in need of a tune and should be better next run(once I apply these maps)?
13340



13341


Thanks again for all your help Joe. Hope I'm not bugging you too much as I try to get this down.:pokey:

joecar
June 11th, 2012, 04:36 AM
Sorry, I have been away during the weekend, I will look at this today... no worries :)


Hmmm, I'm not sure what the lean spike is, how often do you see it.

truethinker
June 11th, 2012, 07:33 AM
No problem man, you're aloud to go away from the forum on the weekend. haha

That spike was only there once or twice over the drive.

I ended up with an overall difference of 23% on my MAF table. Was surprised to see that much of a change. So now I'll go do another log and see if my VE number come in a little higher.

joecar
June 11th, 2012, 12:52 PM
...
I ended up with an overall difference of 23% on my MAF table. Was surprised to see that much of a change. So now I'll go do another log and see if my VE number come in a little higher.
Looks like the lower MAF is close; the upper part looks like it will be ok ( your old MAF doesn't look quite right ), just follow the trend all the way up (manually fill in those 2 holes).

VE table is also close, fill in those 5 holes also.

truethinker
June 23rd, 2012, 12:24 PM
hey Joe, I seem to have everything dialed in pretty close. I re-enabled my MAF and set my PE setting back to normal and activated Lean Cruise and everything seemed to work pretty well. my LTFT's settled in to +/- 0.08% mostly staying at 00. I've noticed though that my WBO2 reading is always off from one side to the other. In terms of Lambda, bank 1 is always 0.010-0.020 more rich than bank two. No matter the condition. Even when lean cruise is active and there is no O2 adjustment being made. Any suggestions on why this may be?

Also, SELBENS and VET won't look right with the MAF re-enabled and other VET setting returned to normal, correct?

Thanks


13407

13408

joecar
June 25th, 2012, 01:46 AM
Since lean cruise is commanding an AFR away from stoich, then then Close Loop trimming cannot be done (since it trims to stoich).

joecar
June 25th, 2012, 01:48 AM
...

Also, SELBENS and VET won't look right with the MAF re-enabled and other VET setting returned to normal, correct?

Sorry, please explain the context...

( MAF is corrected with MAF enabled )

truethinker
June 25th, 2012, 02:11 AM
Hmm I thought the trims were active up until lean cruise kicked in.

Aren't we essentially deactivating Maf based correction by changing the threshold for Maf calculation to 400rpms?

joecar
June 25th, 2012, 02:22 AM
Setting the dynamic air threshold to 400 disables VE... so your MAF is enabled;

SELBEN corrects the MAF, and VET replaces the VE (VET is calculated from corrected MAF);

when you set the dynamic air threshold back to stock (4000 rpm) you will see SELBEN and VET be the same (since MAF is corrected, and VET is calculated from corrected MAF)...

you will also see trims be close to zero (as you have seen).

truethinker
June 25th, 2012, 02:22 AM
Ive just realized that my serial input is backwards. Lambda for bank 2 is actually bank 1. This matches my non-lean cruise fuel trims. Once bank 1 ltft shows a positive value the two wbo2's start showing a closer lambda value.

truethinker
June 25th, 2012, 02:25 AM
Ahh okay, that makes sense. I'm still curious as to why one side would stay at such a different lambda.

joecar
June 25th, 2012, 02:25 AM
I hope you had LC disabled while doing the Calc.VET MAF/VE tuning...?

truethinker
June 25th, 2012, 02:28 AM
I hope you had LC disabled while doing the Calc.VET MAF/VE tuning...?

Yes sir. I only just enabled it.

truethinker
July 2nd, 2012, 11:56 AM
Joe, two quick questions.

First of all, I just wanted to bring up again my question about the difference between my two exhaust banks. My lambda values are different by as much as 0.030lambda with one bank always being leaner than the other. At times it seems that the LTFT's reflect this difference and attempt to correct it(successfully at times) without LC enabled. My average over the whole log is generally around 0.028 lambda difference without LTFT's; 0.011 with LTFT's enabled. I'm just wondering if this may be something I can correct in the tune or if it is a mechanical issue such as a weak injector or something. WBO2 sensors are in the factory locations.

Secondly, my lambda value is switching between 0.970-0.990. What can I adjust to bring this up to around 1.000?

Thanks!

joecar
July 2nd, 2012, 12:11 PM
Differences in measured lambda between banks can be due to many physical things (it's not in the tune):

for example, engine is not symmetrical left-to-right or front-to-rear, each cylinder burns a little differently, each header runner is a different lenngth;

the 0.030 difference you're seeing in Lambda is small (e.g. compare 1.030 with 1.000), it's 3%;

when LTFT's are enabled, the PCM trims each bank to stoich (lambda 1.000) so you won't see as big a difference;

if in OL (LTFT disabled) you're seeing lambda 0.970-0.990, you would make B3605 or B3647 leaner by 0.010-0.030 (which is a very small amount, you might now overshoot).

truethinker
July 2nd, 2012, 01:41 PM
if in OL (LTFT disabled) you're seeing lambda 0.970-0.990, you would make B3605 or B3647 leaner by 0.010-0.030 (which is a very small amount, you might now overshoot).

To expand: This is also the case in closed loop. Even with the trims adjusting it tends to stay in the afore mentioned range if just a little bit closer to 1.000.

joecar
July 3rd, 2012, 12:36 PM
The PCM uses the same tables to calculate air/fuel for each bank...

it uses separate trims to trim each bank...

so any differences between banks can be attributed directly to differences in the O2 sensors, or differences in exuast/intake betweeen the banks, or cylinder-to-cylinder differences.


Did you try swapping the front O2 sensors across banks...?

truethinker
July 3rd, 2012, 12:43 PM
no, I didn't consider that it may be the WBO2's. I'll have to do that.

I wonder if there is something weird going on with one of the LC-1 units since they are simulating the NBO2's? I'll try swapping the WBO2 sensors first and then I'll swap the LC-1 brain if I have to.

truethinker
July 7th, 2012, 12:33 AM
I didn't swap my sensors yet but I was just looking at my most recent log with both lean cruise and dfco disabled and I'm seeing more of those weird maxed out blips on my serial LC-1 reading. It seems to effect both banks and the mV read-out from the simulated narrowband is effected as well. This seems really strange to me but I don't have any LC-1 experience to work from so I don't know. Please review this log when you get a change.

Noted areas are as follows:
Frame: 1373 (both serial banks spike ridiculously lean @ 7.3 lambda and 15.5 lambda. Not sure of any effect on the mV reading)
Frame: 7048 (both serial banks spike ridiculously lean @ 17.5 lambda and 32.5 lambda. Very obviously reflected in the mV reading)
Frame: 9230 (both serial banks spike ridiculously lean @ 7.9 lambda and 23.4 lambda. Not sure of any effect on the mV reading)
Frame: 12161 (One serial bank goes lean and one goes rich @ 10.0 lambda and 0.00 lambda. Not sure of any effect on the mV reading)

Maybe this isn't as strange a thing as I think it is but I figured it's worth looking into.

As a side note, charting the mV read-outs for the two simulated narrowbands gives a good visualization of how one bank is almost always more lean than the other.

13473
13474

joecar
July 8th, 2012, 05:35 PM
The thing to do might be to drill/weld the extra bungs into the headers so you can run both the NBO2's and the WBO2's and see what happens.

truethinker
July 9th, 2012, 02:06 PM
hmm, yea I guess I may have to consider doing that. It's strange though, I don't feel the spike while driving. Also, just the fact that sometimes one side goes lean and one goes rich while other times they both go lean, it seems like it has to be a problem with the lc-1's. Maybe I'll go register at the Innovate forum and post something over there.

joecar
July 9th, 2012, 02:11 PM
Let us know what Innovate says.